Ming Yang fba8906930

[perf] Use direct copy (broadcast) instead of cat for k_nope/k_pe in MLA prefill (#29710) ...

Signed-off-by: Ming Yang <minos.future@gmail.com>

2025-12-11 08:20:45 +00:00

..

mla

[perf] Use direct copy (broadcast) instead of cat for k_nope/k_pe in MLA prefill (#29710)

2025-12-11 08:20:45 +00:00

__init__.py

…

cpu_attn.py

[CPU] Support for Whisper (#30062)

2025-12-10 04:58:42 -08:00

flash_attn.py

[Attention][UX][1/N] Add AttentionConfig and change attention env vars to CLI arguments (#26315)

2025-12-05 09:48:43 -08:00

flashinfer.py

[bug] Fix "Current vLLM config is not set." warnings when FlashInfer attention is used (#30241)

2025-12-10 11:18:51 -08:00

flex_attention.py

[v1] Add PrefixLM support to FlexAttention backend (#27938)

2025-12-07 15:51:36 +00:00

gdn_attn.py

[Attention] Make seq_lens_cpu optional in CommonAttentionMetadata to enable true async spec-decode (#29624)

2025-12-09 17:18:10 -08:00

linear_attn.py

[Attention] add _cudagraph_support for linear attention (#28934)

2025-11-25 12:25:20 +08:00

mamba1_attn.py

[Attention][CUDAGraph] Remove CG padding from attention backends (#29352)

2025-12-02 13:48:08 -05:00

mamba2_attn.py

[Attention][CUDAGraph] Remove CG padding from attention backends (#29352)

2025-12-02 13:48:08 -05:00

mamba_attn.py

[Core] Refactor padding logic and pad for CUDA graphs before attention metadata building (#28579)

2025-11-26 14:07:13 -05:00

pallas.py

…

rocm_aiter_fa.py

[ROCm][Attention] Sliding window support for AiterFlashAttentionBackend (#29234)

2025-11-30 11:31:50 +00:00

rocm_aiter_unified_attn.py

…

rocm_attn.py

[Attention][UX][1/N] Add AttentionConfig and change attention env vars to CLI arguments (#26315)

2025-12-05 09:48:43 -08:00

short_conv_attn.py

[Attention][CUDAGraph] Remove CG padding from attention backends (#29352)

2025-12-02 13:48:08 -05:00

tree_attn.py

…

triton_attn.py

[Attention][UX][1/N] Add AttentionConfig and change attention env vars to CLI arguments (#26315)

2025-12-05 09:48:43 -08:00

utils.py

[Attention] Make seq_lens_cpu optional in CommonAttentionMetadata to enable true async spec-decode (#29624)

2025-12-09 17:18:10 -08:00